|Home | About | Journals | Submit | Contact Us | Français|
Stroke is a leading cause of long-term disability and is the third most common cause of death in the United States and western countries. Twenty percent of strokes are thought to arise from the carotid artery. Histopathological studies have suggested that plaque disruption is a key factor in the etiology of carotid-related ischemic events. Features associated with plaque disruption include intraplaque hemorrhage, large necrotic cores with thin overlying fibrous caps, plaque neovasculature and inflammatory cell infiltrate. In vivo high-spatial resolution, multi-contrast weighted magnetic resonance imaging (MRI) has been extensively evaluated using histology as the gold standard, and has documented reliability in the identification of these key carotid plaque features. In this pictorial essay, we illustrate the capability of MRI for identifying features of disruption-prone and disrupted atherosclerotic carotid plaques.
Stroke is a leading cause of long-term disability and is the third most common cause of death in many countries1. Twenty percent of strokes are thought to be related to extracranial carotid atherosclerosis2. As a means to prevent such cerebrovascular events, carotid endarterectomy and carotid stenting have been advocated in patients with high-grade carotid stenosis. However, the Asymptomatic Carotid Atherosclerosis Study (ACAS) demonstrated that carotid endarterectomy reduced risk for ipsilateral stroke by only 5.9% at five years when compared to best medical management3. Therefore, additional criteria, other than the degree of stenosis, have been sought to better identify patients most at risk of complications from carotid disease.
Based on analysis of histological findings in carotid endarterectomy specimens, plaque disruption is believed to be a major factor in the etiology of carotid-territory ischemic events. Features such as intraplaque hemorrhage, large necrotic cores with thin overlying fibrous caps, plaque neovasculature and inflammatory cell infiltrate4–7 may predispose the atherosclerotic lesion to disruption. Hence, these features represent targets for imaging techniques aimed at identifying high-risk, disruption-prone plaque.
Equally important is the identification of luminal surface disruption, such as fibrous cap rupture, ulceration, and calcified nodules. Retrospective studies have shown that these so- called “culprit lesion” features are associated with a prior history of recent transient ischemic attack or stroke8. Furthermore, they may pose a persistent increased risk for secondary events, as suggested by findings from histology studies demonstrating evidence of repeated cap ruptures,9 and by findings from the North American Symptomatic Carotid Endarterectomy Trial (NASCET). In a subgroup analysis comparing individuals with and without ulcerated carotid plaques, those with ulcers on angiography had a 1.2- to 3.4-fold increased risk for stroke10.
As a non invasive imaging modality with the capability to distinguish tissue characteristics, magnetic resonance imaging (MRI) is an optimal method to characterize the morphology and composition atherosclerotic carotid plaques8,11–28. Multiple centers have shown that MRI can reliably identify fibrous cap status, the lipid-rich necrotic core, intraplaque hemorrhage, and vascular wall inflammation, using histology as the gold standard14–21,23,29–36. Additional advantages of MRI include image generation without ionizing radiation or the need for invasive procedures, which make it an ideal tool for serial, longitudinal study of plaque progression or regression.
Recently published clinical studies demonstrate the potential prognostic value of MRI in patients with moderate carotid stenosis. In a prospective study of 154 patients with 50% to 79% carotid stenosis who were asymptomatic at the time of enrollment, participants underwent baseline carotid MRI and were contacted every three months to identify symptoms of new-onset transient ischemic attack or stroke 25. Twelve cerebrovascular events occurred ipsilateral to the index carotid artery over a mean follow-up period of 38.2 months. Cox regression analysis demonstrated significant associations between ischemic events and presence of a thin or ruptured fibrous cap (hazard ratio, 17.0; P<0.001), intraplaque hemorrhage (hazard ratio, 5.2; P=0.005), and larger mean necrotic core area (hazard ratio for 10 mm2 increase, 1.6; P=0.01) in the carotid plaque. In another prospective study of 64 recently symptomatic patients with 30–69% carotid stenosis, baseline carotid MRI scans were performed to identify intraplaque hemorrhage, and subjects were followed for the development of subsequent transient ischemic attack or stroke37. Thirty-nine (61%) of the ipsilateral arteries demonstrated intraplaque hemorrhage on baseline MRI. Fourteen ipsilateral ischemic events were observed during follow-up. Thirteen of the 14 events occurred ipsilateral to carotid arteries with intraplaque hemorrhage (hazard ratio= 9.8, 95% CI 1.3,75.1, P=0.03). These studies suggest that carotid MRI may provide additional diagnostic criteria to identify patients with moderate carotid stenosis who are at increased risk for subsequent stroke. While these initial results are highly promising, larger multicenter studies are needed to confirm the role of carotid plaque imaging in routine clinical practice.
In this pictorial essay, we illustrate the capability of MRI for the identification of the disruption-prone and disrupted carotid plaque.
The greatest strength of MRI for characterizing atherosclerotic plaque is the availability of multi-contrast weighted protocols utilizing bright- and black-blood techniques. In the past, most applications of carotid MRI have been limited to the evaluation of stenosis using bright-blood MRA. These angiographic pulse sequences produce strong attenuation of signals from stationary tissues, limiting their usefulness for direct imaging of the atherosclerotic plaque. Nevertheless, bright blood MRA using a 3D time-of-flight (TOF) technique presents specific contrast features, which can be helpful in identifying certain plaque components when used in combination with black blood imaging16. Black blood sequences rely on the elimination of the signal from flowing blood and represent a general approach for characterizing the vessel wall, where precise identification of the lumen–wall interface plays a critical role in assessment of morphology and tissue composition of the atherosclerotic plaque26 (Table 1). Implementation of this multi-contrast weighted MRI protocol has been technically successful in 76%–90% of cases. Most of the failures are due to poor image quality secondary to patient motion.15–18.
Numerous studies have demonstrated that combined intensity information from different contrast weightings (Table 1) can be used to identify all major plaque components, including fibrous tissue, lipid-rich necrotic core, calcification, and intraplaque hemorrhage8,11–28. Using established guidelines for the relative intensities of these features to guide manual outlining, studies have shown that quantitative characterization of plaque composition has excellent correlation with histological measurements14,15,17,18,23 and is highly reproducible15. Furthermore, robust automatic classifiers, such as the morphology-enhanced probabilistic plaque segmentations (MEPPS) algorithm27,28 have been developed that automate plaque compositional measurements and achieve similar levels of accuracy and reproducibility. This automated segmentation is based on the fact that various tissue contents such as lipid, calcium, loose matrix, and fibrous tissue have different signal characteristics on each MR weighting 28. The system first determines the probability that each MRI/pixel belongs to each of the 4 tissue types (lipid, calcification, loose matrix, and fibrous tissue). Afterwards, it uses competing active contours38 to identify the boundaries of the high-probability regions for each tissue type.
Studies of advanced atherosclerotic lesions suggest that the thickness of the fibrous cap overlying the necrotic core distinguishes stable lesions from those at high risk for disruption and thromboembolic events8,16. The combined use of bright- and black-blood techniques can aid in the identification of both the fibrous cap and vessel wall components such as the lipid-rich necrotic core (Figure 1). Bright blood time-of flight (TOF) allows for the visualization of the vascular lumen and is capable of identifying the status of fibrous cap16. Following the administration of gadolinium-DTPA, contrast enhanced T1W (CE-T1) differentiates the fibrous cap from the underlying lipid-rich necrotic core and thus allows for the quantification of both components18,21–24. Accurate quantification of both fibrous cap and lipid-rich necrotic core is crucial for measurement of lesion progression.
The spatial resolution of MRI is lower than the proposed histological definition of a thin cap18. However, prospective studies suggest that when an intact fibrous cap is visualized by MRI (categorically defined as a “thick” cap), these individuals have a significantly lower risk for future transient ischemic attack or stroke, compared to individuals where the cap is not visualized (categorized as a “thin” cap) or where the cap is disrupted25. Thus, cap visualization by MRI may have prognostic value. Quantification of minimal cap thickness, and identification of finer structures within the diseased arterial wall will require higher spatial resolution. Improvement in spatial resolution is possible with further technical development, such as higher field strength scanner and the use of 8-element phased array carotid coils39,40.
Intraplaque hemorrhage is frequently observed in carotid atherosclerosis and is purported to arise from plaque neovasculature (Figure 2). Microvessels can be fragile due to lack of support by smooth muscle cells and focal discontinuity of the endothelial lining41. Kolodgie et al 42 suggested that intraplaque hemorrhage may represent a potent atherogenic stimulus by contributing to the deposition of free cholesterol, macrophage infiltration, and enlargement of the necrotic core in a histopathological study of coronary artery specimens. Immunohistochemical staining with the antibody to glycophorin A, a protein specific to erythrocyte membranes, was strongly associated with the size of the necrotic core and degree of macrophage infiltration. The group also noted that rabbit lesions with induced intramural hemorrhage had significantly greater lipid content than control lesions without hemorrhage. Other investigators have noted that erythrocyte membranes contain more free cholesterol than any other cell in the body43.
High-resolution, multicontrast MRI can accurately detect the presence and age of carotid intraplaque hemorrhage13,17,20. Prospective MRI studies have demonstrated that hemorrhage into the carotid atherosclerotic plaque is associated with rapid increase in plaque burden and lipid-rich necrotic core size, and decrease in lumen volume (Figure 3). Furthermore, lesions that had intraplaque hemorrhage at baseline had a greater probability of repeated intraplaque hemorrhage at 18 months13. Rapid progression and the destabilization of plaque driven by the intraplaque hemorrhage can set the stage for subsequent plaque rupture44 (Figure 4).
Inflammation has been described to be an important component of unstable atherosclerotic plaque, whether in the coronary or the carotid, because it contributes to necrotic core size, plaque angiogenesis, and thinning of the fibrous cap through the release of potent matrix metalloproteinases (MMPs)45,46. Regions of the plaque with high-density macrophages are reported to contain MMPs; some of MMPs are co-localized with cleaved collagen46,47. Therefore, recent emphasis has been placed on macrophages as the primary component of rupture.
Contrast enhancement in carotid atherosclerotic plaques has been observed in recent MRI investigations after the injection of a gadolinium-based contrast agent18,21–24,36,48 (Figure 5). Strong contrast enhancement suggests the presence of a vascular supply to the plaque and increased endothelial permeability that facilitates the entry of the contrast agent from the blood plasma. Because neovasculature growth into the plaque and increased endothelial permeability are associated with plaque inflammation, plaque enhancement has been considered a sign of inflammation 36. Since plaque inflammation may have multiple effects that weaken plaque structural integrity, contrast–enhanced MRI may be a tool for detecting plaque inflammation prior to fibrous cap disruption.
Dynamic MRI of contrast enhancement permits quantitative analysis of contrast media kinetics. Uptake is characterized by a parameter for fractional plasma volume (vp) and by a transfer constant (Ktrans) that reflects blood supply, vessel permeability, and the extracellular space23,36. Because these factors are all modulated by the inflammatory process, kinetic modeling has been suggested as a means for characterizing the effects of plaque inflammation. Integrated area under the enhancement versus time curve has also been used to characterize enhancement dynamics in plaque49.
Studies involving dynamic MRI in carotid atherosclerotic plaques have shown that vp correlates strongly with histologically determined neovasculature content 23, a key portal for the entry of inflammatory cells into the atherosclerotic plaques. In subsequent studies, Ktrans has emerged as the preferred marker of plaque inflammation. In a study of 27 patients with histologic results, Ktrans correlated with macrophage (r = 0.75, P < .001), neovasculature (r = 0.71, P < .001), and loose matrix (r = 0.50, P = .01) content36. In another study of 20 subjects, Ktrans was associated with serum markers of inflammation and pro-inflammatory risk factors including high levels of C-reactive protein, low HDL, and smoking48.
However, there is overestimation of neovasculature as measured by dynamic MRI versus histology. This overestimation suggests the presence of interstitial volumes undergoing very rapid exchange of the contrast agent with the blood plasma. Any such regions that come to equilibrium within one time frame of the dynamic sequence will be indistinguishable from blood and therefore included in vp. Development of advanced techniques that also quantify this rapid exchange may lend further insight into the kinetics of contrast agents in plaque.
Another contrast agent used in plaque imaging employs ultrasmall particles of superparamagnetic iron oxide (USPIOs), which have been shown to accumulate in macrophages within atherosclerotic plaques and to lead to characteristic losses in signal intensity on MR images50. Superparamagnetic iron oxides, however, require multiple imaging sessions over periods of 24 hours or more51.
It is well known that high shear stress can destabilize the vascular wall (Figures 6a,b). Shear stress acting on the vessel wall plays an important role in many processes in the cardiovascular system primarily focused on the regulation of vessel lumen and wall dimensions. There is ample evidence that atherosclerotic plaques are generated at low shear stress regions in the cardiovascular system52. In addition, plaque rupture has been more frequently observed at the proximal, upstream side of the site of maximal stenosis, which is exposed to higher wall shear stress (WSS)53. It is also hypothesized that high WSS at the upstream side of the plaque has a biological effect on the fibrous cap by inducing antiproliferative activity by endothelial cells and therefore may enhance plaque vulnerability 52–55.
In order to quantify WSS, computational fluid dynamics (CFD) models are built with realistic boundary conditions. Imaging techniques such as MRI provide information regarding vascular geometry and inflow conditions. Figure 7 illustrates a CFD model of an atherosclerotic internal carotid artery. After generating the computational mesh, the shear stress field acting on the luminal surface is computed from the velocity field. Although this approach is commonly accepted 54, it is still unclear ho variability in geometry reconstruction and restricting assumptions on blood rheology or vessel wall compliance may affect the accuracy of the results.
Phase-contrast MRI, however, is a promising, noninvasive technique for determining various in vivo blood flow characteristics. WSS of the carotid artery can be assessed semi automatically with good to excellent reproducibility without inter- or intraobserver variability using model-based segmentation of phase-contrast MRI by determination off low volume and maximum flow velocity in cross-sections of these vessels56. However, WSS is site specific, a finding which opposes the notion that physiological WSS values are maintained at a constant magnitude in all parts of the arterial system. Among the WSS values obtained at the same site by different investigators there is qualitative agreement. However, differences exist in absolute values mainly due to the dependence on the method used to obtain WSS values from velocity data57,58. Large variations in absolute WSS levels are also reported within one species and between species59.
Rupture of the fibrous cap, with the resultant exposure of thrombogenic subendothelial plaque constituents, is believed to be the critical event that leads to thromboembolic complications in atherosclerotic carotid artery disease5–7. Histopathology studies have shown that the prevalence of carotid plaque rupture was significantly higher among patients with a past history of an ischemic neurological event5–7,60. Disrupted plaques in patients affected by stroke were characterized by the presence of a more severe inflammatory infiltrate, compared with that observed in the transient ischemic attack and asymptomatic groups6.
High-resolution MRI with a 3-dimensional TOF protocol is capable of distinguishing intact, thick fibrous caps from intact thin and disrupted caps in atherosclerotic human carotid arteries in vivo8,16 (Figure 8). With the addition of contrast-enhanced T1-weighted sequence MR can more readily identify the intact, thick fibrous cap18 (Figures 1,,4).4). Identification of fibrous cap rupture with MRI is highly associated with recent transient ischemic attack or stroke8.
Studies of carotid endarterectomy specimens have shown thatplaque ulceration and thrombosis are more prevalent in symptomatic patients5–7. Ulceration is more common in symptomatic patients regardless of side of carotid symptoms, whereas thrombus is associated with ipsilateral symptoms and plaque ulceration5. A thrombotically active carotid plaque associated with high inflammatory infiltrate was observed in 71 (74.0%) of 96 patients with ipsilateral major stroke compared with 32 (35.2%) of 91 patients with transient ischemic attack (P < .001) or 12 (14.6%) of 82 patients who were without symptoms in a study6.
Case studies have shown promise in the detection of carotid atherosclerotic ulceration using multisequence cross-sectional MRI44,61 (Figures 9,,10).10). Adding longitudinal black-blood MR angiography to multisequence high-spatial-resolution cross-sectional MR images can increase the ability of MRI to identify carotid plaque ulceration62 (Figure 11).
Calcified nodules were first reported in acutely thrombosed coronary arteries, but have since been reported in carotid arteries. Surface nodules can have exposed thrombogenic surfaces or can be encapsulated63. Transverse MR images demonstrate the usefulness of TOF in the multisequence MR protocol for detecting nodules that present with the same hypointense signal as the background blood in black blood sequences64 (Figure 12).
This review demonstrates the value of high-spatial resolution, multi-contrast weighted MRI techniques in the identification of the disruption-prone and disrupted carotid atherosclerotic plaque. Based on these histologically verified techniques, we highlight plaque features that are associated with rapid progression, surface disruption, and an increased risk for subsequent ischemic events. These features include the necrotic core, intraplaque hemorrhage, the thin and ruptured fibrous cap, ulceration, and calcified nodules. In addition, MRI lends itself to collecti ng information on wall sheer stresses that may influence both rupture of the fibrous cap and plaque progression.
We expect that new and continuing advances in MRI technology such as higher field strength, dedicated phased-array coils for higher signal-noise ratio and larger coverage of carotid artery, multi slice motion-sensitized driven-equilibrium (MSDE) turbo spin-echo (TSE) sequences to improve suppression of plaque-mimicking artifacts65, and 3D isotropic sequences for better luminal surface and plaque delineation66, will provide even more tools to better characterize the vulnerable atherosclerotic carotid plaque.
No conflict of interest:
Daniel S. Hippe
Marina S. Ferguson
Chun Yuan – RO1 HL56874
Thomas Hatsukami – clinical research grant from Hoffmann-LaRoche
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.